An Acoustic Based Local Positioning System for Dynamic UAV in GPS-Denied Environments

Abstract

Highlights A positioning system for mobile UAVs is needed in a GPS-denied environment. A Doppler shift compensation algorithm was proposed. The successful detection rate and positioning errors were 90.2% and 75.9 ± 39.1 mm, respectively. The system performance meets the requirements for mobile UAVs in an indoor environment. Abstract. An alternative positioning system, the Spread Spectrum Sound-based Local Positioning System (SSSLPS), was developed to extend the usage of UAVs in GPS-denied environments, such as greenhouses. To date, it has been proven that the SSSLPS could provide a centimeter-level locational accuracy of static objects and dynamic ground robots. In this research, the ranging and positioning performance of SSSLPS were evaluated with the system mounted on a UAV. Meanwhile, to minimize the Time of Arrival (ToA) estimation error and improve positioning accuracy, a novel Doppler shift compensation algorithm was proposed. Experiments were conducted, and the measurement accuracy was evaluated by comparing the measurement of the motion capture system. The results show a successful detection rate of 90.2% using the proposed algorithm and a position error of 75.9 mm. These results demonstrate that the SSSLPS with spectrum peak algorithm can improve positioning accuracy and enable the SSSLPS to be an accurate positioning system for UAVs in a greenhouse. Keywords: Dopper shift, Greenhouse robots, Positioning system, ToA estimation, UAV.